simpledirect/Vinci-MLE-8B-1.0
Text Generation • 9B • Updated • 263
Vinci MLE
Vinci MLE — Models specialized for machine learning engineering, with transparent evaluation results and deployment options across model sizes
Note Full bf16 weights, 17.6 GB across four shards. Per-file digests in SHA256SUMS and PROVENANCE.json.
simpledirect/Vinci-MLE-30B-1.0
Text Generation • 29B • Updated • 130Note Full bf16 weights, 57.7 GB across twelve shards. Per-file digests in SHA256SUMS and PROVENANCE.json.
simpledirect/Vinci-MLE-123B-1.0
Text Generation • 125B • Updated • 162Note Full BF16 weights, 250.1 GB across 27 shards. Digests in SHA256SUMS; Mistral AI Modified MIT licence.
simpledirect/Vinci-MLE-8B-1.0-GGUF
Text Generation • 9B • Updated • 121Note GGUF conversions: F16, Q8_0, Q5_K_M, Q4_K_M. Digests published; tiers not capability-benchmarked.
simpledirect/Vinci-MLE-30B-1.0-GGUF
Text Generation • 29B • Updated • 114Note GGUF conversions: F16, Q8_0, Q5_K_M, Q4_K_M. Digests published; tiers not capability-benchmarked.
simpledirect/Vinci-MLE-123B-1.0-GGUF
Text Generation • 125B • Updated • 71Note GGUF: Q4_K_M (74.9 GB), Q5_K_M (88.3 GB), Q8_0 (132.9 GB). Runtime checks documented; Mistral AI Modified MIT licence.